For companies like Microsoft and Mozilla, Safari coming to Windows means that one more competitor will be thrown into the Windows browser battlefield. While neither company has expressed dismay with Apple’s decision to put Safari on Windows, on Monday executives from both Microsoft and Mozilla expressed a lack of concern for their new (Windows) foe.
… somebody should tell them that the browser war is passe.
The browser war is over; it’s the “standards” war now. Apollo v Flash v Silverlight fighting for the closed web. ODF v OOXML, PDF v XPS, MP3 v DRM, HD-DVD v BluRay.
As for Safari, Apple are being blinded by their iTunes halo. Safari does not behave like a proper Windows app, and generally people don’t like software that doesn’t fit in. It looks freaky, you can’t close the window when maximized by clicking the corner of the screen, you can only resize on one corner, it uses its own fonts and anti-aliasing, there’s no split between the titlebar and the menu. It’s an awful program, with an excellent rendering engine.
i think the fonts actually look good.
not only all that but it has memory leaks galore. I was running meebo in it the other day just to see how well it ran it and it sucked 75mb of ram and was climbing and was slow to respond to anything before I killed the process. Safari sucks on windows. at least the initial version.
For comparison, firefox 2.0 on the same machine had 10 tabs open and was using 80something megs
not only all that but it has memory leaks galore. I was running meebo in it the other day just to see how well it ran it and it sucked 75mb of ram and was climbing and was slow to respond to anything before I killed the process
I always hear this yet I don’t experience it. Currently my instance of Firefox has been open for less than an hour, has 5 tabs open and is sucking 122 MB of RAM. I’m a web developer, and usually have 10 – 15 tabs open and have to restart firefox once or twice per day due to excessive (> 500MB) RAM usage/slowness.
Safari (on OS X) tends to use less RAM, but locks up more often (and doesn’t include the tools I need, hello firebug!)
Anyone else experiencing this?
Safari (on OS X) tends to use less RAM, but locks up more often (and doesn’t include the tools I need, hello firebug!)
Anyone else experiencing this?
Well, I do also experience the need for Firefox when doing webapp work
. But I also do experience less burning of my lap when I use Safari compared to Firefox on my Macbook. And there are a couple of sites I frequent that cause Firefox 1.5 & 2.x to require a restart whereas Safari is okay — Firefox suddenly can’t scroll, can’t close the tab, can’t close the window, etc. Oddly Firefox on Windows&Linux can render the sites in question just fine.
It’s a shame, as Safari sucks, but it seemingly sucks less than all other browsers on the Mac (for me, at least!).
Yeah, whats up with that? – screw flash, silverlight, air/flex – use that for embedded content, not for whole sites!!!
Windows web developers can now easily test their sites against Safari. That has to be a good thing.
For one, Windows Safari does not act any close to Mac’s safari, period. Even if Apple fixes dozens of critical problems it still probably has in a matter of time, since 2 OS have 2 different set of fonts and who knows when a Windows upgrade might screw up Safari and there’s no guarantee that both Safari look and act the same, there’s just the fact that serious web developers have to test ‘both’ Windows and Mac Safari, now that is more work than less work…
Test your site on IE6, IE7, Firefox on Mac/Win/Linux, Opera on Mac/Win/Linux and finally Safari on Mac/Win, and your web site should look and behave mostly fine.
Safari on Windows uses the same fonts and anti-aliasing system as Mac safari, as well as the same engine. If anything, it has the most accurate rendering between platforms of any browser, including Firefox, which does not share the same font rendering on Windows/Mac.
Joel Spolsky: http://www.joelonsoftware.com/items/2007/06/12.html
Dave Shea: http://mezzoblue.com/archives/2007/06/12/a_subpixel_s/
Jeff Atwood: http://www.codinghorror.com/blog/archives/000885.html
In a nutshell, Apple is attempting to bring their philosophy about fonts/typefaces over to the Windows realm in hopes of wooing potential converts over to the Mac way of doing things.
My opinion? Simple: 9 out of 10 folks will prefer whatever they are more used to if a blind test were to be done on what they think looks better. Most people don’t really care unless they are heavy into graphics/desktop publishing.
Edited 2007-06-13 23:09
It seems rather obvious that font rendering is broken in some way or another.
I seem to remember OS X having much better fonts than this beta version of Safari.
“I seem to remember OS X having much better fonts than this beta version of Safari.”
The point of my comment was that fonts will appear “broken” to users of the other platform, i.e. fonts on Windows look strange to longtime Mac users who are looking at them for the first time, and vice versa.
Mac fonts are optimized for printing on media (WYSIWYG on the screen to printed media) whereas Windows fonts are optimized for screen readability at the expense of printed media looking somewhat different. I recommend reading the links above as they go in depth to the actual pixel level as to why this is true.
That being said, the amount of programming that Apple has to do to get Mac-like font rendering on Windows is more than likely an immense amount of effort as they won’t be able to use any of the high level rasterizing Win32 API’s and will have to roll their own. Kudos to Apple for that, but IMO virtualization is easier, cleaner, and requires less effort. And if this is indeed geared towards iPhone developers being able to test on a Windows machine, why not just release an iPhone emulator for Windows? That would be geared more towards developers in a neater package…it’s what I’d prefer, but perhaps I’m in the minority?
It just seems like an awful lot of effort for not much ROI.
See this screenshot of the Windows & Mac version side by side. The fonts arn’t broken at all, they simply look out of place on Windows because it’s different than Cleartype
http://img522.imageshack.us/my.php?image=picture1ip3.png
That was what I was going to say, I think the fonts look nice on OSX, but Safari looks like crap because the fonts are so different than every other font on the screen.
apple is a microsoft company.
there will be two microsoft browser for windows. great!
i’m not using windows, but one of my friends tested it on windows and safari dies on apple.com. i know it’s beta, but that’s just funny!
“apple is a microsoft company.
there will be two microsoft browser for windows. great!”
I’m surprised you haven’t been modded down for this troll.
Microsoft bought, what, 2% of Apple stock in 1997? Two Microsoft browsers indeed – what rubbish!
true enough, microsoft has nothing to do with apple’s products, they just invested to save face. they bought no patents, collaborate on none of the core products each other are working on (exceptions MIGHT be bootcamp, and in the application ports each company has devised) to say apple is a microsoft company is unfounded and completely rediculous. Thats like screaming Apple is my company because I own some shares.
“apple is a microsoft company.
there will be two microsoft browser for windows. great!”
What? Just do some reading, MS and Apple are two very different companies, with different target markets, different design philosophies, and different business practices. Research is important.
I’ve seen several reports of Safari taking upwards of 600-700 MB of memory after a few hours of use. Yes, I know it’s a beta on a platform that Apple isn’t used to writing software for. But that’s still a ton of memory, and is a large issue they need to get sorted out before they are to be taken seriously.
As a web developer I’m pleased that I’ll no longer have to test for Safari on a Mac, assuming the rendering engines on both platforms are identical. That’s about the extent of my ‘jubilation’ though…it looks terrible on Windows as it sticks out like a sore thumb and doesn’t follow Windows design guidelines.
>>I’ve seen several reports of Safari taking upwards of 600-700 MB of memory after a few hours of use.
I’ve heard numerous reports of Firefox doing the same thing. RAM usage is excessive in most Windows and Linux programs these days.
> a platform that Apple isn’t used to writing software for
Counting iTunes and QuickTime not enough experience to write another Windows application sounds like, they never keep a reference they build up to their next project to me.
What they did is… rushed for WWDC. Nothing more. Too bad they refrained to call it alpha or developer reference version but beta.
iTunes was successful because of it was the first program for Windows that could interface with the iPod. Remember, iTunes for Windows was released very shortly after the iPod became successful. iTunes was good enough that people with an iPod began to use it for playing, buying, and burning songs. Some people without an iPod also began to use it.
I don’t think Apple is that stupid to introduce Safari to Windows for these reasons:
1) Because they think their browser is good enough that Windows users will adopt it (as specified by Kroc)
2) Because it’s beneficial for web designers (as suggested by Tyr. and jayson.knight)
Instead, I suggest Apple will introduce a new service that can only be accessed by Safari. Apple wants this service to be accessible to Windows users as well.
Edited 2007-06-13 23:42
The whole point of Apple developing and releasing a version of Safari for Windows: the iPhone. That’s the absolute largest reason for them to do so. Now more sites and Web 2.0 apps will work with the iPhone because developers will have a much easier time checking cross-browser compatibility.
The second largest: revenue sharing from Google et al via the search bar a la Mozilla.
That’s it.
I believe they are targeting all web developers. That’s why they went through all the trouble of replicating the Mac OS font rendering. With Safari available for Windows it’ll be much easier to make sites compatible with Safari and the Mac.
Especially if it handles SSL sites correctly on all computers: due to some weird race condition within the bowels of FF code, more often than not, it fails in some weird timeout or “The document contains no data” on my machine (dual p3-450) regardless of whether I’m using it in Windows 2k3 or BeOS 5.03.
IIRC, for a free browser that should handle SSL sites correctly, I’d have to put up with ads (Opera) or I have to deal with all the issues of IE on my current system, since at least it works consistently.
So, Apple, please, make a boringly predictable Windows version!
>> I’d have to put up with ads (Opera)
Uhm, Opera hasn’t been Adware for almost a year and a half now (September ’05 – Opera 8.5), has been adblock capable via user.js, user css since that time, and with version 9.x now has a full blown block content function built in.
And under BeOS your last Opera version was 3.6, and that was ad free as well.
Though if you want to see why you are having SSL issues in the bloated pig known as firefox, one doesn’t have to look very far.
>> my machine (dual p3-450)
You might want to consider running hardware that is less than a decade old.
Before you make such silly statements and insults about someone’s hardware or where they spend their money, at least back them up with facts: the pentium-3 is from (at earliest) early 1999. Besides, the hardware is capable of rendering a webpage faster than they’re typically available for download, so why is more processing power needed, when there’s enough RAM? The thought that a web browser would need several Ghz worth of processor clock shows how much of pigs they’ve become, or what people have come to expect. I have no problem watching mpeg videos on this machine.
Besides, developing on such a machine helps keep coding for efficiency in mind, to some degree, even though my past employment has been developing for embedded hardware with much less oomph. IE runs perfectly fine on the processor, and I have enough enough RAM: why can’t FireFox also run decently? And the speed of the processors is completely irrelevant to the race condition: properly written multithreaded software works correctly, fast or slow, while incorrectly written multithreaded software is unsafe at any speed: if it never fails on another machine with the same input conditions, then that’s just dumb luck, and the curse of “But it runs on MY machine!” which is often the hallmark of careless developers that don’t test.
Oh, and this hardware has survived 100F + days without AC (it is rare in northwest Washington to have that weather or AC), running perfectly stable, a direct lightning strike, and other things, since commissioned in 1999. How many Pentium 4 machines or AMD’s can claim such things, all without overheating your apartment or house? Yeah, by today’s standards, it’s slow, but it runs most everything I need, is perfectly stable, is BeOS compatible (most modern hardware won’t boot BeOS) etc.
>> the pentium-3 is from (at earliest) early 1999
ooh, I was off by a year… BFD.
>> Besides, the hardware is capable of rendering a webpage faster than they’re typically available for download, so why is more processing power needed, when there’s enough RAM? The thought that a web browser would need several Ghz worth of processor clock shows how much of pigs they’ve become, or what people have come to expect.
That I can agree to, to a certain extent. The problem is how long can you backwards code support for processors that lack opcodes your entire current product line has – which is part of where Apple’s problem lies. The lowest intel they’ve ever sold is a core solo (even the Celeron M in the AppleTV is a Yonah 430 – that’s basically a mobile core solo), their software reflects that.
>> why can’t FireFox also run decently?
Because it’s a fat bloated pig – moreso in BeOS? (let’s face it, XUL and pervasive multithreading mix like oil and water) Not like you have a choice there – ooh, Opera 3.6 and whatever the built in one is called. (I use it so rarely)
>> most modern hardware won’t boot BeOS
Well, if you know where to get the right boot disks, it will. I’m able to run 5.03 just fine on a A64 4000+ – it’s a matter of patching into this century. (oh, sorry, this 8 years)… Not that I have any reason to anymore, or ever really did given that nothing I need to do work is available for it. BeOS, the best OS nobody wrote REAL applications for. (can’t even use it for my media work – no soundfont or full on EMU APS support)
>> How many Pentium 4 machines or AMD’s can claim such things, all without overheating your apartment or house?
Depends on how much of a wussy the owner is about fans really. I’ve fallen in love with xClio cases for their 10″ side fans… Which is why my 8800GTS and A64 4000+ runs 10C lower than the spec says.
Though for AMD’s, yeah… heats a bitch, but if you are REALLY worried about heat on a modern system, that’s what socket M is for… or even better a VIA. I’m willing to bet a 1.5ghz Via C7 will own your dual processor on both speed and heat – there’s a reason you can run them without fans.
But if it does what you need it to do, more power to you – just don’t be shocked when the rest of the world leaves you behind. Web developers have all but abandoned supporting Netscape 4, and are looking to drop IE 5.x (I already have) – contemporaries of your hardware. Software designers are not far behind… and this is nothing new. Remember when MMX came out? MMX only on software? 386 only? 286 only? Ringing any bells? New processors with new opcodes come out yearly, and it is only a matter of time before processors lacking those opcodes are left out in the cold. If making the program run faster on TODAYS hardware means leaving stuff that’s more than five years old run like molassas or worse, leave them out completely, that’s historically what happens.
I actually agree with a LOT of what you said, but after doing this for three decades (oh, I’m sorry, 29 years) I can honestly say (and I tell my customers this) if you do not upgrade your hardware and software every three years, FIVE YEARS MAX, do NOT be surprised when new software doesn’t work, new peripherals you might want to add don’t work or have limited functionality, and the computing world in general leaves you behind. If it didn’t, I’d still be using my Trash-80 Model 1.
…and remember, for every working machine more than five years old, there are ten dead ones on the trash heap.
Edited 2007-06-14 11:05
…and remember, for every working machine more than five years old, there are ten dead ones on the trash heap..
We need to get off this rollercoaster of planned obsolescence.
>> why can’t FireFox also run decently?
>Because it’s a fat bloated pig – moreso in BeOS? (let’s
> face it, XUL and pervasive multithreading mix like oil
> and water) Not like you have a choice there – ooh,
> Opera 3.6 and whatever the built in one is called. (I
> use it so rarely)
FYI, you are out of your water here. The windows version is much more bloated, when it comes to lines of code, not object-size. It has special handling for disk-access lots of of hidden windows to run message pumps and has the same problem of syncing threads as all other platforms.
It actually helps BeOS that it is multithreaded. It’s a pain in the ass handling syncing either way, multithreaded or not. BeOS just has some nice things to cope with it. Anyone writing an acceptable browser will not manage to keep it single-threaded anyway.
I do agree that Firefox is bloated though. (I’m actually not a big fan of it at all)
/TQH, Firefox developer for BeOS/Haiku
>> It actually helps BeOS that it is multithreaded.
>> It’s a pain in the ass handling syncing either way,
>> multithreaded or not. BeOS just has some nice
>> things to cope with it. Anyone writing an
>> acceptable browser will not manage to keep it
>> single-threaded anyway.
Then why does Firefox for BeOS (5.03, Max, leaked Dano and Yellowtab) run like molassas in february on the same hardware compared to Linux or Windows?
Sorry, my results are having a disconnect from what you said. I thought since FF seems to damn near run it’s own threading/tasking and memory management, running that atop a pervasive threading system would be more of a headache… as evidenced by the piss poor performance of said browser on that platform. (hell, running IE5.2 under MacOS7 in Basalisk II under BeOS is faster)
Or are the builds we’ve seen so far just really badly done – and the ones being done for Haiku way better? (Please say yes)
Edited 2007-06-15 09:28
Because our builds are not like the release-builds you’ve used on those platforms. Their release-builds are statically linked builds WITH NO debug-symbols. Our release-builds are statically linked WITH debug-symbols. It almost doubles the size of executable and therefore it’s performance, but it allows our small team to track and fix bugs, which is highest priority. For the same reason we usually don’t include optimization-flags to the compiler, this tend to vary though.
Also important is that they use gcc4, we use gcc2.95.3. You can probably find some figures on the net how they compare.
Third, the network-stacks under BeOS and Zeta are probably not comparable to Linux and Windows. I know it’s very true for net-server, but you could always compare ftp-speed on the platforms.
In fact release-builds perform better in BeOS than other platforms, built with comparable flags on my machine using javascript-rendering tests.
I installed it here, and… It’s less than alpha quality. I can’t type text, be it in the location bar, in the search bar, or inside the web page. I can’t change the web page either, the tool bar doesn’t show up, there’s no tab, no keyboard shortcut work, and as soon as you type a key, the browser crashes. RAM usage is very high also, which is strange, being a KHTML browser. OS integration is terrible, it looks like iTune and Quick Time, being an OS X application (obviously!). The only page that it displays is the Apple.com web site but text is not rendered (only images). When they fix all these bugs, it will be interesting for webdevs to test their sites in Safari (not everyone has a Mac). But I don’t expect many people to use it as a default browser.
I think it is interesting that a rendering engine from KDE Desktop, of Linux origins, is now a Windows browser with fine tuning supplied by Apple. The rendering speed is faster than Firefox but the GUI doesn’t offer anything compelling for Windows users. I don’t see anything that would make a Windows user currently using IE7 or Firefox want to switch to Safari. I like the idea that the Safari browser might become a way to get more Apple software onto Windows.
Edited 2007-06-14 01:17
>> but the GUI doesn’t offer anything compelling
>> for Windows users.
Which is why as I’ve said a few other places it’s like a trip in the wayback machine to netscape 4 or IE 3 in terms of the UI.
Where are we going today Mr. Peabody?
1997 Sherman. Redmond, Washington.
I know I’m spoiled by Opera’s “Over the top” customizability, but Safari isn’t even as configurable in terms of the UI as IE4 (windows) was. (though the same thing could be said about IE7)
It makes a lot of sense though that the UI would be so ‘nuetered’ – It’s made by apple folk for apple folk, and for them gecko based browsers are ‘feature heavy’ out of the box… remember, until three years ago their primary choice of browser was IE 5.2 – if you can find a browser with a more ‘locked in’ UI, you are probably running Lynx. When you compare to IE 5.2, the UI damned near identical in functionality and customizability.
BUT at least under windows it lets you maximize it – unlike the epileptic crack addict routine the maximize button gives you in MacOS. “Maximize? Ok, we’ll make it a little larger… kind of… Why would you want it to run full screen anyways?”
Edited 2007-06-14 01:50
BUT at least under windows it lets you maximize it – unlike the epileptic crack addict routine the maximize button gives you in MacOS. “Maximize? Ok, we’ll make it a little larger… kind of… Why would you want it to run full screen anyways?”
That’s because OSX does not have a maximize button. The closest way to even call it a maximize button is to call it an efficient maximize button. It does not maximize the window to use the whole screen but, it maximizes the window the most size that is *needed* but the window.
If you do a lot of multitasking with different documents and windows, this makes much more sense than having a button that maximizes the window to take up 100% of your useable screen space. It’s still a personal preference thing but I much prefer OSX’s zoom button to any other maximize button out there.
>> That’s because OSX does not have a maximize
>> button. The closest way to even call it a maximize
>> button is to call it an efficient maximize button.
>> It does not maximize the window to use the whole
>> screen but, it maximizes the window the most size
>> that is *needed* but the window.
Which is NOT something I would ever want or need, since if I’m making it larger it’s because I’m either going to be manipulating the zoom in my browser, want the extra border around the content in photoshop because I don’t feel like sitting their trying to select an edge pixel, or in general want more screen space than MacOS ever seems to think I’d want. It’s a total /FAIL/ on usability for me leading me to taking more steps to do something really simple… and never QUITE actually getting what I want.
>> If you do a lot of multitasking with different
>> documents and windows, this makes much more sense
>> than having a button that maximizes the window to
>> take up 100% of your useable screen space.
Because showing a little tiny corner immediately tells me which of those twelve open .c, .h, .php, .html and .css files I’m working on it is… Just like expose being useless because zoomed down to crappy little thumbnails those files all look SO different. (which is why flip3d, beryl and expose can suck my… for **** sake just give me a **** LIST) There’s a reason my G4 Powerbook with OSX is relegated to ‘cute toy’ and my A64 4000+ desktop is where REAL work gets done.
See, I have this thing called a taskbar that LISTS all running programs not just by program name, but the filename that’s being manipulated too… Of course that I run that on my right display in portrait mode expediates matters even more… that I’m able to run each of the browsers in the center and left DISPLAYS even more so. The dock is a tinkertoy compared to the taskbar in Windows, KDE, Gnome, XFCE – hell, let’s just say ‘RoW’ (rest of world).
Besides, how is adding more than one display to a new Mac that costs under $2500 working out these days?
Edited 2007-06-15 04:45
Unofficially, all Macs in the past few years have worked with a second display (extended desktop, not mirrored). My old iBook G4 was used with a second monitor for a long time.
Officially, all Intel Macs, as well as PowerBooks, support it.
Edited 2007-06-15 07:41
The only reason Apple is even releasing this rough beast right now is because web devs can use it to test their iPhone apps. Secondly, the more users that use it is further justification for sites to support it. Obviously everyone testing the beta helps them with a more polished product, too.
I’m not sure how many people will be actually “using” it, but thank god one person hit the nail on the head. It was even stated by apple that it was released so people on Windows could work on iphone “apps.” I believe that the version of safari on the phone is pretty much the full monty.
iPhone sales.
It’s a trojan, just like iTunes was for iPod sales.
Simple.
I struggle to see why Apple has done this.
I am a huge Apple fan, in fact, I’m using Safari (3) on the Mac right now. Because I’m such an Apple fan I grabbed the Win version too to have a look.
The thing is, even when this is scrubbed up and running well, not eating memory and the bookmarks work correctly on the Win version without crashing etc., I can’t see why Apple did this.
As one reader suggested, maybe Apple will introduce a Safari only service that they want Windows users in on. In that case, I hope a Linux version is introduced soon too.
The other reason, and probably closer to the truth is getting Windows developers out there developing for the iPhone too. As this renders Apple fonts and has access to Apple services, this would be a good reason. As before, if this is the case, then I hope a Linux version comes out soon (which of course it wont).
I honestly don’t think it will gain much market share, not unless there is some really compelling reason. Firefox is still my weapon of choice right now on the PC (and that may change), and even that only has around 15% or whatever it is.
Oh, there are some nice features that Safari has, like resizing this window I am typing in and the find feature is very nice too, but not mind blowing or anything… It does render pretty fast too, but for me, if this window renders in 0.48 seconds instead of 1.03 seconds will hardly make my day any better.
Leopard is going to have some really nice features, but as it stands right now, Safari isn’t one of them yet…
There’s many things I’m not happy about with Firefox, but what keeps me on it are the add-ons. For the geek/power-user crowd that’s the big compelling reason. For mom and pop, it’ll be IE. Apple is going to have something compelling for it to be relevant.
See, the only reason I don’t use firefox is because of the addons. If the only reason the browser is useful is because of it’s extensibility, then the browser is flawed in my opinion.
That and I think it is a memory hog, i find it unstable, i don’t like the path the project is taking in terms of priorities.
If the only reason the browser is useful is because of it’s extensibility, then the browser is flawed in my opinion.
I would agree, but personally feel that Firefox is a decent browser even without the addons. It seems I’m only using 6 of them now, and the only one I really need is Adblock Plus.
That and I think it is a memory hog,
Yes it is.
i find it unstable,
I always hear complaints about that, but haven’t experienced any problems myself. Back in the 1.0 days I had a few crashes due to the Acrobat plugin, but haven’t had more than 3 or 4 crashes in total since they fixed it – about the same as Windows.
i don’t like the path the project is taking in terms of priorities.
I’m a little skeptical of this as well, but I’m taking a wait and see approach. I’m looking forward to having Konqueror work in Windows, but for now I’m sticking with Firefox.
Edited 2007-06-14 08:51
This app may rule on OSX, but nobody in the Windows world is going to take it seriously. It may be good for testing sites on the iPhone, but that’s about the only purpose it’ll serve. Mac apps on Windows tend to be absolutely horrible, and based on what I’ve read here and elsewhere, Safari doesn’t look like it’s going to be an exception.
I dont know bout that, ive been using it but I dont find it that great compared to firefox, but for my other officemates who doesnt customize anything, they love it and they use it as their browser.
The Mac version of Safari appears to be a Cocoa application (isn’t it? I don’t have a copy to check right now.)
What is the Windows version? Has anyone identified the language/framework? It seems like a lot of trouble to duplicate with a different toolkit, and you’d have a fundamentally different product. Could this mean that Cocoa for Windows actually exists and is usable for a fair sized project?
If so, that’s far and away the most interesting thing to come out of WWDC since the Intel transition. Sooner or later they will let it out….
Cocoa is AFAIK the current name of the OpenStep / NeXT APIs.
The OpenStep APIs were ported to Windows under the name “OPENSTEP Enterprise” (OSE). I wouldn’t be surprised if some parts of this continued to be maintained even after the Apple buyout.
The only major parts missing would be the Aqua UI, but that work was to some extent already done for iTunes/Quicktime and the Core* frameworks (although iTunes/Quicktime probably have some them to a limited extent).
So it is highly likely that “Cocoa for Windows” does exist to some extent, but is probably not complete enough to allow trivial porting of OS X apps to Windows.
Safari isn’t a trivial app, and it likely uses some Apple-specific technology, suggesting that they’ve ported at least part of their private frameworks too. Lack of those private frameworks is a major problem for GNUStep ports.
On the other hand, with access to all the source, they may have ported only what they needed, leaving lots of holes in the API that would be a hassle for others.
It’s hard to believe, though, that they’d go to all that trouble without some thought of releasing it. It’s not like it wouldn’t *sell* on Windows, even as a barebones porting system using gcc. Cocoa makes most Windows app frameworks look like bad jokes, and while I loathe and despise XCode and IB, they’re no worse than Visual Studio.
I’ve not installed Safari on my Windows machine, but I read this at scobleizer.com (Robert Scoble’s blog):
http://scobleizer.com/2007/06/11/welcome-to-the-blurry-but-fast-bro…
“#
Actually, Apple has accomplished more than just releasing a fast browser, with significantly better standards support than IE, on Windows.
This isn’t a simply a “port” of Safari to Windows, but the same Objective-C source code used in Mac OS X. Essentially, it’s a full blown Cocoa application running on Windows. The included runtime DLLs are the fabled Yellow Box for Windows.
This would be the equivalent of Microsoft delivering a beta .NET application, such as Expression Studio, for Mac OS X.
Comment by Scott — June 11, 2007 @ 6:45 pm”
When NeXT was still around, NextStep (based on OpenStep) ran on NT and BSD; when Apple bought them, OSX NextStep was branded “Cocoa” and NT NextStep was branded “YellowBox” (which did see a limited release, IIRC). (The exact history escapes me, but you get the gist.)
So it appears that Apple is using the old YellowBox dlls, which you can think of as NT Cocoa, and therefore Win Safari is in essence a “Cocoa” app.
Note that later comments in that same Scoble blog speculate that this may be the cause for Win Safari’s huge memory requirements. I’ve see reports of Win Safari using 300-400MB for only 5 or six tabs, and someone above mentioned reports of it using 700MB.
Oh how I wish it was “partner” or “colleague” instead of “foe”. If we continue with the same war mindset then we’ll never have a standards based web and the mess that it is now will continue to dominate. A browser is just an application for displaying a page written in a well-documented programming language. Just think what would be if we had different C compilers each treating C code “differently” than the rest. So why do we have to suffer from browser incompatibilities…
there will be two microsoft browser for windows. great!
ipod speaker
http://www.mp4-converter.net/dvd-to-mp4/
And today, Apple released Safari 3.0.1 for Windows to fix those issues (supposedly). Any takers on this yet?
All of my instability experiences have come from the mac side of things. I don’t tend to use linux on the desktop and the only thing I use windows for is WOW and Eve online.
my macbook can’t play wow worth crap
Now you don’t need a mac to test website for your mac customers!
Too bad it’s not linux version.
oh no, safari on windows – it is awful. http://www.prevedgame.ru/in.php?id=20508